An Online Packing Heuristic for the Three-Dimensional Container Loading Problem in Dynamic Environments and the Physical Internet

Ha, Chi Trung; Nguyễn, Trung Thành; Bui, Lam Thu; Wang, Ran

doi:10.1007/978-3-319-55792-2_10

Cited by 23 publications

(18 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Training on a spatial resolution of 100 × 100 takes about 12 hours and a single decision time is less than 10 ms. We compare our method with some state-of-the-art online bin packing methods to demonstrate the superiority. OnlineBPH [17] and OnlineDRL [50] are two representatives, where OnlineBPH is non-learning based and OnlineDRL learns how to pack from the data. [50] also proposes a heuristic baseline called boundary rule method.…”

Section: Methodsmentioning

confidence: 99%

“…Similar strategies have also been adapted to Online BPP works like [17,20,46,50]. Different from the offline setting, the size information of coming items is unknown for the agent to optimize the packing and the packing order can not be adjusted as well.…”

Section: Related Workmentioning

confidence: 99%

“…It makes Online BPP a much more challenging problem. Some works [17,46] have employed the hand-coded heuristics based on the human experience. Meanwhile, works like [50] have adopted deep reinforcement learning to learn how to pack things effectively through trials and error optimization.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Learning Practically Feasible Policies for Online 3D Bin Packing

Zhao¹,

Zhu²,

Xu³

et al. 2021

Preprint

View full text Add to dashboard Cite

We tackle the Online 3D Bin Packing Problem, a challenging yet practically useful variant of the classical Bin Packing Problem. In this problem, the items are delivered to the agent without informing the full sequence information. Agent must directly pack these items into the target bin stably without changing their arrival order, and no further adjustment is permitted. Online 3D-BPP can be naturally formulated as Markov Decision Process (MDP). We adopt deep reinforcement learning, in particular, the on-policy actorcritic framework, to solve this MDP with constrained action space. To learn a practically feasible packing policy, we propose three critical designs. First, we propose an online analysis of packing stability based on a novel stacking tree. It attains a high analysis accuracy while reducing the computational complexity from O(N 2 ) to O(N log N ), making it especially suited for RL training. Second, we propose a decoupled packing policy learning for different dimensions of placement which enables high-resolution spatial discretization and hence high packing precision. Third, we introduce a reward function that dictates the robot to place items in a far-to-near order and therefore simplifies the collision avoidance in movement planning of the robotic arm. Furthermore, we provide a comprehensive discussion on several key implemental issues. The extensive evaluation demonstrates that our learned policy outperforms the state-of-the-art methods significantly and is practically usable for real-world applications.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Learning Practically Feasible Policies for Online 3D Bin Packing

Zhao¹,

Zhu²,

Xu³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The previous studies successfully implemented online packing processes. Some have employed the fine-tuned heuristics based on the product packing routine [ 16 , 17 , 18 ] and the others have actively used deep reinforcement learning for their methods to learn how to pack things effectively themselves through trials and errors [ 11 , 19 , 20 , 21 ]. However, due to the high computational load of online optimization or learning complexity, objects have been assumed to be boxes without rotations or 3D problem has been decomposed into multiple 2D problems by packing objects layer by layer.…”

Section: Related Workmentioning

confidence: 99%

Smart Pack: Online Autonomous Object-Packing System Using RGB-D Sensor Data

Hong

Kim

Lee

2020

Sensors

View full text Add to dashboard Cite

This paper proposes a novel online object-packing system which can measure the dimensions of every incoming object and calculate its desired position in a given container. Existing object-packing systems have the limitations of requiring the exact information of objects in advance or assuming them as boxes. Thus, this paper is mainly focused on the following two points: (1) Real-time calculation of the dimensions and orientation of an object; (2) Online optimization of the object’s position in a container. The dimensions and orientation of the object are obtained using an RGB-D sensor when the object is picked by a manipulator and moved over a certain position. The optimal position of the object is calculated by recognizing the container’s available space using another RGB-D sensor and minimizing the cost function that is formulated by the available space information and the optimization criteria inspired by the way people place things. The experimental results show that the proposed system successfully places the incoming various shaped objects in their proper positions.

show abstract

“…In such settings, for an inexpensive practical robotic automation, it is mostly assumed that the previously settled items cannot be readjusted, and a new incoming package cannot be put under the previous packages since such locations are usually difficult and expensive for robots to accesswhich we refer to as practicalrobotic constraint. An algorithm based on the empty maximal space approach has been proposed in [8] to solve an online 3D BPP for rectangular shapes. The empty maximal space strategy may lead to a highly sub-optimal solution.…”

Section: Introductionmentioning

confidence: 99%

Efficient Algorithm Based on Adaptive Window–Model Predictive Control for Automatic Stacking in Warehouse Center

Chang

Sebghati

2021

IEEE Access

View full text Add to dashboard Cite

In this article, three-dimensional (3D) stacking problem, as a mandatory task in the warehouse distribution centers, is smartly represented by a new mathematical formulation. A model predictive control (MPC) scheme is proposed to optimally solve the problem. The MPC cost function consists of several sequential criteria to suitably put the incoming packages on the dedicated cart and fully use up the cart capacity as much as possible. The practical and hard constraints are taken into consideration without confronting with a complicated optimization problem. A theorem is presented to guarantee the constraints satisfaction. Regarding the burden of computations, the proposed strategy is efficiently applicable to large carts by introducing the adaptive window (AW) idea. The efficacy of the proposed AW-MPC framework is shown by a numerical case study. INDEX TERMSAdaptive window, matrix representation, model predictive control, stacking problem.

show abstract

An Online Packing Heuristic for the Three-Dimensional Container Loading Problem in Dynamic Environments and the Physical Internet

Abstract: The version presented here may differ from the published version or from the version of the record. Please see the repository URL above for details on accessing the published version and note that access may require a subscription.

Cited by 23 publications

References 30 publications

Learning Practically Feasible Policies for Online 3D Bin Packing

Learning Practically Feasible Policies for Online 3D Bin Packing

Smart Pack: Online Autonomous Object-Packing System Using RGB-D Sensor Data

Efficient Algorithm Based on Adaptive Window–Model Predictive Control for Automatic Stacking in Warehouse Center

Contact Info

Product

Resources

About